Claude Code cheat sheet

A scannable Claude Code reference: 25 short snippets across 11 topics, each linking back to the lesson it came from.

At a glance

Quick snippets

Installing and configuring Claude Code

Install and authenticate

# install the CLI
npm install -g @anthropic-ai/claude-code
claude --version

# run it from the project root, not from your home directory
cd ~/work/my-service
claude
# first launch: sign in, then verify the working directory it reports

Instructions and permissions

<!-- CLAUDE.md, at the repository root, committed with the code -->
# Working agreement

- Run `npm test` before claiming any task is complete; paste the summary.
- Never edit files under dist/ or any generated file.
- Prefer the existing helpers in src/lib/ over new dependencies.
- Two-space indentation, no semicolons, single quotes.

<!-- These are instructions, not enforcement. Permissions decide what can run. -->

Instructions and permissions

{
  "permissions": {
    "allow": ["Bash(npm test)", "Bash(npm run lint)", "Read(./src/**)"],
    "deny":  ["Read(./.env)", "Read(./secrets/**)", "Bash(curl:*)"],
    "ask":   ["Bash(git push:*)", "Bash(npm install:*)"]
  }
}

Full lesson: Installing and configuring Claude Code →

A typical editing workflow

Explore and plan before editing

> /init
  reads the repository and writes a starter CLAUDE.md you can edit

> how does the auth middleware attach a user to the request? cite the files.
  < explanation with file:line references

> that is wrong about the refresh path - it re-reads the cookie at line 42.
  now propose a plan to return 401 when the token is expired, and change nothing yet

> proceed, then run the tests. keep the change to src/auth/ and its test file.

Edit, verify, commit

git status --short        # clean before the agent starts
git switch -c agent/expiry-401

# ... one focused request, then review
git diff                  # read the whole patch
npm test                  # run the gate yourself
git add -p && git commit -m "Return 401 for expired tokens in auth middleware"

# clear context between unrelated tasks so old noise stops influencing edits
# (inside the session: /clear)

Full lesson: A typical editing workflow →

Permissions and safety with agents

Modes and rules

claude --permission-mode plan

# A narrow allow-list beats a broad mode: allow the commands you actually use
#   "allow": ["Bash(npm test)", "Bash(npm run lint)", "Bash(git diff:*)", "Read(./src/**)"]
#   "deny":  ["Read(./.env)", "Bash(rm -rf:*)", "Bash(git push --force:*)"]

Full lesson: Permissions and safety with agents →

Managing the context window and session hygiene

What fills the window

# inside a session
/context        # token usage broken down by source
/cost           # what this session has spent so far
/clear          # start over with an empty conversation
/compact        # summarise the conversation so far and continue

# from the shell
claude -c                       # continue the most recent session
claude -r                       # pick a past session to resume
claude --fork-session -c        # resume, but branch instead of overwriting

Session hygiene that keeps quality up

# a cheap way to cap how much output enters the window
npm test 2>&1 | tail -40

# check what a file costs before reading it
wc -l src/server/*.js

# non-interactive, with a hard ceiling on turns
claude -p "list the exported functions in src/api" --max-turns 3

Resuming, branching and long-running work

# a long refactor split into three resumable sessions
claude -p "rename the User model to Account across src/models and update imports" --max-turns 20
claude -c -p "now update the tests to match the rename"
claude -c -p "summarise what changed and which files still reference the old name"

Full lesson: Managing the context window and session hygiene →

Slash commands and reusable prompts

A command is a markdown file

# after saving as .claude/commands/review.md
/review error handling in the upload path

# subdirectories create a namespace
# .claude/commands/db/migrate.md  ->  /db:migrate

Arguments and shell substitution

---
description: Explain a failing test file
allowed-tools: Read, Bash(npm test:*)
---
The test file $1 is failing. Run it, then explain the failure.

Recent commits touching it:
!git log --oneline -5 -- $1

Keeping a command library useful

claude
> /help            # commands appear alongside the built-ins
> /review          # project command, from .claude/commands/review.md

# personal shorthand, available in every repo
mkdir -p ~/.claude/commands

Full lesson: Slash commands and reusable prompts →

Scoped instructions: memory, rules and path-specific context

Path-scoped rules

---
paths:
  - "src/api/**/*.ts"
  - "src/workers/**/*.ts"
---

# Server-side rules

- Every handler validates input with the schema from src/schemas.
- Never log a full request body; log the request id instead.
- Return the error envelope: { error: { code, message } }.
- Database access goes through the repository module, never raw SQL here.

Why mid-session edits do not apply

# the honest sequence
vim CLAUDE.md          # add: "always run npm test before finishing"
# ... the running session does not know about this

# a new session picks it up
exit
claude

Full lesson: Scoped instructions: memory, rules and path-specific context →

Subagents and context isolation

Invoking and designing them

# list and manage definitions
/agents

# delegate explicitly
> use the code-reviewer subagent on the staged diff

# fan out deliberately: several independent searches in parallel
> use three subagents to locate the rate limiter, the retry policy and the timeout config

Concurrency and cost

---
name: migration-auditor
description: Audit a directory for patterns that block a framework migration.
tools: Read, Grep, Glob
model: haiku
---

Audit the path given in the task. Report only files that need manual changes,
each as: path, line, pattern found, suggested replacement.
Do not explain the migration. Do not read test fixtures.

Full lesson: Subagents and context isolation →

Hooks and deterministic automation

Blocking a call

{
  "hooks": {
    "PreToolUse": [
      {
        "matcher": "Edit|Write|MultiEdit",
        "hooks": [
          { "type": "command", "command": ".claude/hooks/guard-secrets.sh" }
        ]
      }
    ]
  }
}

Hook safety and debuggability

# verify that stdin really is what you think it is
echo '{"tool_name":"Write","tool_input":{"file_path":".env","content":"A=1"}}' \
  | .claude/hooks/guard-secrets.sh ; echo "exit=$?"

# then force the whole run to fail when the guard is wrong
claude -p "append DEBUG=1 to .env" --allowedTools "Edit" ; echo "exit=$?"

Full lesson: Hooks and deterministic automation →

Skills, plugins and marketplaces

Bundling a setup as a plugin

my-plugin/
  .claude-plugin/
    plugin.json          # name, version, description, author
  commands/
    review.md            # becomes /review
  agents/
    code-reviewer.md     # a subagent definition
  skills/
    release-notes/
      SKILL.md
  hooks/
    hooks.json           # hook configuration shipped with the plugin

Bundling a setup as a plugin

{
  "name": "acme-standards",
  "version": "1.2.0",
  "description": "Review commands, release-notes skill and lint hooks for Acme services"
}

Full lesson: Skills, plugins and marketplaces →

Cost, model choice and troubleshooting

Model choice and effort

# per session
/model                 # switch interactively
/cost                  # usage for this session

# per invocation
claude -p "rename getUser to fetchUser everywhere" --model haiku
claude --model opus

# from the environment
export ANTHROPIC_MODEL=claude-sonnet-4-5

Diagnosing loops and thrashing

# give evidence, not opinions
npm test 2>&1 | tail -60 > /tmp/fail.txt
claude -p "Here is the real failure output. Read /tmp/fail.txt and the handler it names, then state the root cause before changing anything."

# cap the blast radius of an experiment
claude -p "make the failing test in src/cart.test.ts pass without touching any other file" \
  --max-turns 12

Health checks and bad edits

claude doctor        # installation, auth and environment health
claude --version
/status             # account, model and workspace in a session
/doctor             # same checks from inside a session

Full lesson: Cost, model choice and troubleshooting →

Building your own agent with the Agent SDK

When to build rather than drive the CLI

npm install @anthropic-ai/claude-agent-sdk
# or, for Python
pip install claude-agent-sdk

Custom tools and evaluation

const cases = [
  { q: "Why did order A-1042 fail?", expect: ["declined", "insufficient funds"] },
  { q: "Which orders are stuck in review?", expect: ["A-1030"] },
];

for (const c of cases) {
  const answer = await runOnce(c.q);
  const ok = c.expect.some((e) => answer.text.toLowerCase().includes(e));
  console.log(ok ? "pass" : "FAIL", "|", c.q, "| turns:", answer.numTurns);
}

Full lesson: Building your own agent with the Agent SDK →

FAQ

Related cheat sheets

AI Basics AI Agents Math for AI Machine Learning scikit-learn TensorFlow

Last refreshed 2026-09-27.